- observium - lists.observium.org

Re: [Observium] INVALID_MSGID
by Tom Laermans 17 May '14

17 May '14

Hrm, looks like my grep skills failed me this morning. I'll take a look later today. Thanks for looking at this! Simon Smith <simonsmith5521(a)gmail.com> wrote: >Hi All, > >i think i found the bug :-D > >includes/alerting/email.inc.php -> line 93 > >$headers['Message-ID'] = '<' . md5(uniqid(time())) . '@' . $params['localhost'] . '>’; > >$params isn’t set anywhere within the file !? > >anybody have an idea what it should actually be set to? > >regards > >Simon > >On 15 May 2014, at 12:45, Tom Laermans <tom.laermans(a)powersource.cx> wrote: > >> Wut? >> >> On 15/05/2014 08:45, Nikolay Shopik wrote: >>> That's is observium using some loosy script to generate mails >>> >>> On 15.05.2014 3:56, Simon Smith wrote: >>>> Hi, >>>> >>>> i have noticed when the Observium is sending alerts to me, our mailcleaner is blocking them, >>>> >>>> one reason is because the Message-ID is invalid or corrupted when it was sent !? >>>> >>>> for some reason seeing the Message-ID is being seen and sent as >>>> >>>> msgid=<cb0e6df8f4186cc23b4ba07c8bb65af4@> >>>> >>>> i got this information from my sendmail mail.log file and our mailcleaner logs too >>>> >>>> is this a bug in sendmail or the observium? >>>> >>>> regards >>>> >>>> Simon >>>> >>>> _______________________________________________ >>>> observium mailing list >>>> observium(a)observium.org >>>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >>>> >>> _______________________________________________ >>> observium mailing list >>> observium(a)observium.org >>> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium >> >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > >_______________________________________________ >observium mailing list >observium(a)observium.org >http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

2 3

Dell Hardware Sensors
by Travis Kreikemeier 16 May '14

16 May '14

Very little information about this in the Wiki or configuration guide. I would like to get hardware sensors working in Observium for Dell servers, PE 2950, R310, R610, etc. I want to alert when any sensor is not OK. For example, hard drive failure, power supply failure, etc. >From what I think is possible. That information is available in SNMP on a physical Linux server if you setup SNMP SMUX Peer to the Dell OpenManage agent running on the server? Does Observium support it? However, it appears this is not possible on physical Windows? As well, as it look like that hardware SNMP nformation is not available either on ESXi 5+ because of lack of SNMP support integration with Dell OpenManage agent VIB. I'm on correct on this? Or can Observium get this hardware health from WS-MAN on ESXi, or IPMI to iDRAC hardware card? It would be awesome to see a Observium definitive hardware monitoring guide for Dell and then HP. Any information or help would be appreciated! Observium rocks! Oh, and the Alerting system in Pro is awesome! However, little bit of a learning curve. I created 9 alerts that I think should cover most issues. Such as high cpu, low memory, device down, device rebooted, network port at half duplex, network port down, high port utilization, etc. It would have been ideal if there was a Wiki page with all of those just laid out for examples. Everyone just keeps adding more and more example alerts to that page. TRAVIS KREIKEMEIER

3 2

Edit bill ports display bug
by Spencer Gaw 16 May '14

16 May '14

Hey all, I've noticed a display issue on the Edit Bill view, specifically within the Bill Ports section - see attached. The ports display without issue when viewing the bill, but do not appear at all (except for the delete button) in the Edit view. Looks like this issue is present in r5478 and r5492. This isn't a huge deal by any means but if someone could take a look when they have a second, it would be much appreciated. Thanks! Regards, SG

1 0

Incorrect Port VLAN Information
by Casey Dinsmore 16 May '14

16 May '14

I'm testing out Observium and I've noticed a problem with some of the VLAN assignments. Our shop runs HP Procurve switches and we've seen this problem on both the 2824 and the 5308xl. Here is what Observium sees vs the switch config: http://pastebin.com/FGC4z8eJ Is this a known issue? Thanks, Casey

2 2

fisheye.observium.org not working
by Stefano Cordio 16 May '14

16 May '14

Hey guys, I really like follow your work through fisheye.observium.org, but since few days I don't reach it anymore, it shows only a root directory listing of an Apache webserver. Has something changed? Kind Regards, Stefano

2 1

Re: [Observium] Distro Script - CloudLinux
by Adriaan Smuts 16 May '14

16 May '14

Sorry, I forgot to add we use Observium Professional: Observium 0.14.5.5435 From: Adriaan Smuts Sent: Tuesday, May 6, 2014 10:57 AM To: Observium Network Observation System Subject: Distro Script - CloudLinux Hi Guys, Can you please add CloudLinux to the /usr/bin/distro script? # uname -r 2.6.32-458.23.2.lve1.2.52.el6.x86_64 # /usr/bin/distro RedHat 6.5 http://www.cloudlinux.com/ Regards Adriaan Smuts Systems Administrator - Windows [file:///C:\Users\adriaan\AppData\Roaming\Microsoft\Signatures\WebAfrica_Files\signature_logo.jpg] ________________________________ Direct Line +27 21 464 9565 Reception 086 000 9500 Website www.webafrica.co.za<http://www.webafrica.co.za/>

4 19

Gaps in graph data
by Teixeira, Mike 15 May '14

15 May '14

3 4

INVALID_MSGID
by Simon Smith 15 May '14

15 May '14

Hi, i have noticed when the Observium is sending alerts to me, our mailcleaner is blocking them, one reason is because the Message-ID is invalid or corrupted when it was sent !? for some reason seeing the Message-ID is being seen and sent as msgid=<cb0e6df8f4186cc23b4ba07c8bb65af4@> i got this information from my sendmail mail.log file and our mailcleaner logs too is this a bug in sendmail or the observium? regards Simon

3 5

Geocoding shows wrong locations
by Ray, Bjoern Karsten 15 May '14

15 May '14

Hi there, I am running several VLAN in a distributed environment in Germany. I am using the longitude/latitude options to finally not ending up on "An island in the middle of the ocean". Yesterday I added another two VLANs and the map shows devices inside e.g. in India. As I am using private subnet ranges, only, I am wondering why this happens at all and could be done to get over of this little trouble. Subnets I am using are: 192.168.50.0/24 192.168.55.0/24 192.168.60.0/24 192.168.70.0/24 192.168.80.0/24 192.168.90.0/24 192.168.100.0/24 192.168.110.0/24 192.168.120.0/24 ...where the C-octet reflects the VLAN-ID. Thanks in advance!

3 4

Observium Alert Setup
by Simon Smith 15 May '14

15 May '14

Hi, hopying someone can help me, bought the observium, and its brilliant, just what i was looking for and even the graphs and stuff however i can’t seem to get the alerts to work correctly!? i want to monitor a device (not a port) so when it goes offline or does a reboot it will send me an email or run a custom command of my own… is the any guide to help set this up for me? or even if someone could simply tell me how to set it up sorry for being thick, but it just abit confusing to setup a simple alert on a device thats all Regards Simon

2 3

Re: [Observium] ping alerts
by Redish, Graham (PPPO/CONTR) 14 May '14

14 May '14

Tom, Is there a good write up on the new alerting system? I've seen various generic / site specific examples but no documentation that shows me what I can alert on and how to go about setting it up. If you could point me in the right direction or give me some tips I'd appreciate it. >In the new alerting system you can differentiate between snmp is down and ping is down (device_status_type will be either 'ping' or 'snmp'). >Tom >On 13/05/2014 01:09, Adam Armstrong wrote: > I question the existence of a valid scenario where snmpd can't reply > to sysDescr but the server isn't considered broken. > > You're also talking about the /old/ alerting system, which is > deprecated and universally hated and shunned. :> > > adam. > > On 2014-05-12 17:58, Joshua Elson wrote: >> Is there a way to disable alerts for ping when snmpd isn't accessible? >> We get lots of false alerts when servers are under high load. We >> basically just want to be notified if a host is actually down (not >> pinging) and not when snmpd chokes out. >> >> Thanks! >> >> JOSH ELSON | Linux Systems Administrator MAXLINEAR | >> www.maxlinear.com [1] >> 2051 Palomar Airport Road, Suite 100 | Carlsbad, CA 92011 >> >> ------------------------- >> >> This electronic mail (including any attachments) may contain >> information that is privileged, confidential, and/or otherwise >> protected from disclosure to anyone other than its intended >> recipient(s). Any dissemination or use of this electronic mail or its >> contents (including any attachments) by persons other than the >> intended recipient(s) is strictly prohibited. If you have received >> this message in error, please notify us immediately by reply e-mail >> so that we may correct our internal records. Please then delete the >> original message (including any attachments) in its entirety. >> >> Links: >> ------ >> [1] http://www.maxlinear.com/ >> >> _______________________________________________ >> observium mailing list >> observium(a)observium.org >> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium > _______________________________________________ > observium mailing list > observium(a)observium.org > http://postman.memetic.org/cgi-bin/mailman/listinfo/observium ------------------------------ Subject: Digest Footer _______________________________________________ observium mailing list observium(a)observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium ------------------------------ End of observium Digest, Vol 46, Issue 30 *****************************************

4 4

troubleshooting sonicwall reporting failures
by Josh Bierman 14 May '14

14 May '14

Hello, Im having an issue with my sonicwall firewalls. they show 15 minute blank spots in reporting. Primary and fail-over devices, at different times. nothing stands out in the logs, and no one is screaming that internet was down for 15 minutes. Suggestions on getting to the bottom of this? is it just a graphing failure? (see attached screen shot) [image: Inline image 1] Joshua Bierman Director of Technology Yeshivat Noam

1 0

Upgrading to Observium Professional
by Jeremy Alberts 14 May '14

14 May '14

Hallo, I am thinking about upgrading to Observium Professional. Before I do this however, I was wondering if someone could answer a question. At the moment I am using Observium to monitor all my Cisco switches and Windows servers. The thing I miss the most in Observium Open Source edition is not being able to receive alerts on the following: § Alerting Memory at certain thresholds (Windows machines) § alerting CPU at certain thresholds (Windows machines) § Alerting disk I/O at certain thresholds (Windows machines) Now I thing this is possible with the Professional edition. Can someone confirm that this is the case? With kind regards, Jeremy

2 1

ping alerts
by Joshua Elson 13 May '14

13 May '14

Is there a way to disable alerts for ping when snmpd isn't accessible? We get lots of false alerts when servers are under high load. We basically just want to be notified if a host is actually down (not pinging) and not when snmpd chokes out. Thanks! Josh Elson | Linux Systems Administrator MaxLinear | www.maxlinear.com<http://www.maxlinear.com/> 2051 Palomar Airport Road, Suite 100 | Carlsbad, CA 92011 [Description: Description: http://wiki/download/attachments/1704274/Maxlinear+Logo.png?version=1&modif…] ________________________________ This electronic mail (including any attachments) may contain information that is privileged, confidential, and/or otherwise protected from disclosure to anyone other than its intended recipient(s). Any dissemination or use of this electronic mail or its contents (including any attachments) by persons other than the intended recipient(s) is strictly prohibited. If you have received this message in error, please notify us immediately by reply e-mail so that we may correct our internal records. Please then delete the original message (including any attachments) in its entirety.

3 3

Disk i/o statistics
by Fabrice Charlier 12 May '14

12 May '14

Hi all, We are currently moving our observium server from an old community edition (0.13.10.4585) to a new powerful server running Professional edition (0.14.4.5391). On the new one, all disk i/o statistics for server devices don't appear in the web interface. After some research, I found that rrd files for disks i/o are not generated. For example, same server monitored by old instance: http://pastebin.com/gF7RA9bU and on new one: http://pastebin.com/Y9ePqyst Maybe I'm missing some new configuration option? Thanks, Fabrice

4 7

Add device failing
by Robert Williams 12 May '14

12 May '14

Hi, Just tried to add a new device (r5478) but it’s failing with no feedback, even at the command line. root@OBSERVIUM:/opt/observium # php add_device.php server.name.domain Observium v0.14.5.5478 Add Device(s) Try to add server.name.domain: Trying v2c community xyz ... root@OBSERVIUM:/opt/observium# It was ok a couple of days ago and I’m not sure how to try to diagnose this further as there is no feedback either at the command line or the web interface (which just says “Trying v2c community xyx …” the same as the command line does). To rule out a duff device I removed an existing working device and tried to re-add it. Fails with the same error also and I’m unable to re-add that device. Any pointers please? Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

3 4

Calix E7-2 port disabling
by Crozier, Doug 11 May '14

11 May '14

I have 14 Calix E7-2 OLTs in this environment that I want to monitor. The first two I added to Observium(PE 5447) showed up with over 1000 ports each. I'd like to disable polling/alerting on all but a handful of ports on each of these devices. I used the "All" button on the device's Ports tab to get this done, but it only disables the first 333 ports. Same behavior on the second OLT. The remaining 900+ ports are all still enabled for polling. Manually selecting the slider for 50 or so of the remaining ports to disable them fails to take effect after clicking "Save". Is there a way to disable the remaining ports, then enable the few that I am interested in? Thanks Doug

3 3

dpkg agent oddness
by Robert Williams 10 May '14

10 May '14

Hi All, I’ve just added the dpkg agent script for the agent to a server, it already had other agents on it which are all working OK (apache, ntp etc.). Now, when I ./dpkg locally it gives the package list just fine. However, Observium shows the package list with some extra items at the top which read <blank> and “Fatal error:” [cid:image001.png@01CF6C41.93B4FDE0] Now on every poll (manual or scheduled) I also get the event log entry “System Arch -> Directive” [cid:image002.png@01CF6C41.F2135120] This dpkg script is working fine on all the other hosts I run it on, and everything else is working fine on this particular host apart from this element of the dpkg script. If I run the agent module in debug then I notice that at the end of the script I get: Fatal error: Directive 'allow_call_time_pass_reference' is no longer available in PHP in Unknown on line 0 Fatal error: Directive 'allow_call_time_pass_reference' is no longer available in PHP in Unknown on line 0 ) So I guess this is what is being interpreted by the poller as a ‘package’ and the ‘architecture’ of the system. However, I do not get this error when I run it locally, only when it’s run from Observium remotely. So - Any ideas or suggestions what to try next? Target host is Ubuntu 14.04 LTS (3.13.0-24-generic x86_64) and I have several others running 14.04 which are polling fine as far as I’m aware. Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

1 1

Relative times not showing properly...
by Michael Loftis 09 May '14

09 May '14

I'm actually not certain when this started to happen as I've noticed it before, but it seemed like it was just in old events, but now, I've noticed that all relative times are getting computed incorrectly, they'll all show like "364 days, 17h 13m 3s" when you hover for an actual time (system event log particularly) the timestamps are right...it's something going wonk doing date math I'm guessing... 0.14.5.5458 - php 5.4.4 (just svn updated to make sure it wasn't fixed) not sure what else is relevant... -- "Genius might be described as a supreme capacity for getting its possessors into trouble of all kinds." -- Samuel Butler

3 4

weekly/monthly reporting tool?
by Joshua Elson 09 May '14

09 May '14

Hi all, I was wondering if there is any sort of reporting tool for Observium? We are looking to get a weekly and/or monthly break down of CPU and memory for all 100 or so of our servers. It has become tedious to go through each server one by one each time we need to gather all these details. As always, thanks in advance! Josh ________________________________ This electronic mail (including any attachments) may contain information that is privileged, confidential, and/or otherwise protected from disclosure to anyone other than its intended recipient(s). Any dissemination or use of this electronic mail or its contents (including any attachments) by persons other than the intended recipient(s) is strictly prohibited. If you have received this message in error, please notify us immediately by reply e-mail so that we may correct our internal records. Please then delete the original message (including any attachments) in its entirety.

3 4

Ruckus ZoneDirector
by Eduardo Schoedler 09 May '14

09 May '14

Hi, Do you plan to provide the discovery of the Ruckus ZoneDirector? Thanks. Regards, -- Eduardo Schoedler

3 3

Alerts only with alert checker
by Eduardo Schoedler 09 May '14

09 May '14

Hi, We are using Observium Pro, congratulations, it's a great tool! But I'm having trouble with alerts and need your help. I have created an alert check for bgp peers, and is ok. But Observium still want send alerts from himself :). How to disable it? Thanks in advance. Regards, -- Eduardo Schoedler

3 4

Add disk status to Buffalo Terastation
by Stefano Cordio 09 May '14

09 May '14

Hey guys, I'd like to implement an addition for my NAS box, a Buffalo Terastation. This NAS sends a daily report about the available space and the state of the disks. Observium already grabs details about the space, so I'd like to also add support to disk status/diagnostics (I already have the needed MIB files). The following is a MIB snippet: ------------------------------------------------------ nasDiskStatus OBJECT-TYPE SYNTAX INTEGER { notSupport(-1), normal(1), array1(2), array2(3), standby(4), degrade(5), remove(6), standbyRemoved(7), degradeRemoved(8), removeRemoved(9) } MAX-ACCESS read-only STATUS current DESCRIPTION "The state of local disks. This value is read from 'diskx' line in file /etc/melco/diskinfo. notSupport(-1) indicates the disk slot is not exist. normal(1) indicates the disk is not part of RAID array. array1(2) indicates the disk is part of array1. array2(3) indicates the disk is part of array2. standby(4) indicates the disk is standby and power-off. degrade(5) indicates the disk has degraded and removed from array, and power-off. remove(6) indicates the disk has removed from array by user and power-off. standbyRemoved(7) indicates the disk is standby and removed from NAS. degradeRemoved(8) indicates the disk has degraded and removed from array and NAS. removeRemoved(9) indicates the disk has removed from array and NAS." ::= { nasDiskEntry 2 } ------------------------------------------------------ I'm not sure which is the best location in the device page for showing these informations. Could we consider these as a kind of sensor or is it something inventory-related? Actually I need a design advice. Would be really nice if you could also give me a pointer to a similar existing code. I strongly believe in learning by example :-) Thanks, Stefano

2 2

Juniper vlans
by gregorcy 09 May '14

09 May '14

Hi, Our Juniper vlans are not showing up correctly. I am wondering if it is something we are doing or possibly a bug in the Juniper code. Here is an image of what we are seeing: http://i.imgur.com/d0MM7Z8.png Thanks for any help, -- Cheers, Brian Gregorcy IT Manager University of Utah Department of Chemical Engineering 801.585.7170

3 3

Re: [Observium] Observium] weekly/monthly reporting tool?
by Adam Armstrong 08 May '14

08 May '14

We need to add filtering capability to this page so that it can be reduced to only the important entities. Adam. Martin Smith wrote:

1 0